Scholarly culture: How books in adolescence enhance adult literacy, numeracy and technology skills in 31 societies

Abstract

A growing body of evidence supports the contention of scholarly culture theory that immersing children in book-oriented environments benefits their later educational achievement, attainment and occupational standing. These findings have been interpreted as suggesting that book-oriented socialization, indicated by home library size, equips youth with life-long tastes, skills and knowledge. However, to date, this has not been directly assessed. Here, we document advantageous effects of scholarly culture for adult literacy, adult numeracy, and adult technological problem solving. Growing up with home libraries boosts adult skills in these areas beyond the benefits accrued from parental education or own educational or occupational attainment. The effects are loglinear, with greatest returns to the growth in smaller libraries. Our evidence comes from regressions with balanced repeated replicate weights estimated on data from 31 societies which participated in the Programme for the International Assessment of Adult Competencies (PIAAC) between 2011 and 2015.

Full text

1
Scholarly culture: how books in adolescence enhance adult
literacy, numeracy and technology skills in 31 societies
Abstract
A growing body of evidence supports the contention of scholarly culture theory that immersing
children in book-oriented environments benefits their later educational achievement, attainment and
occupational standing. These findings have been interpreted as suggesting that book-oriented
socialization, indicated by home library size, equips youth with life-long tastes, skills and knowledge.
However, to date, this has not been directly assessed. Here, we document advantageous effects of
scholarly culture for adult literacy, adult numeracy, and adult technological problem solving. Growing
up with home libraries boosts adult skills in these areas beyond the benefits accrued from parental
education or own educational or occupational attainment. The effects are loglinear, with greatest
returns to the growth in smaller libraries. Our evidence comes from regressions with balanced
repeated replicate weights estimated on data from 31 societies which participated in the Programme
for the International Assessment of Adult Competencies (PIAAC) between 2011 and 2015.
KEYWORDS: HOME LIBRARIES, BOOKS IN ADOLESCENCE, SCHOLARLY CULTURE , ADULT
LITERACY, NUMERACY AND DIGITAL PROBLEM SOLVING, CULTURAL MOBI LITY
1. Introduction
How does culture enhance life chances? The most prominent theory of this tradition,
Bourdieu’s argument about cultural reproduction, builds on Max Weber's insight into the
importance of culture in boundary maintenance by elite status groups. It posits that elite
families equip their children with “widely shared, high status cultural signals (attitudes,
preferences, formal knowledge, behaviours, goods and credentials) used for cultural
exclusion” (Lamont and Lareau, 1988: 156) thereby securing educational advantages for their
children and reducing opportunities for other children. Critical to this argument are three
elements: (1) the signals are arbitrary – they do not actually enhance educational or
occupational performance; (2) the signals are difficult and/or time consuming to acquire, so
they are difficult to fake; (3) the elite have near-exclusive access to these signals. A host of
empirical studies inspired by this argument conclude that cultural capital, misconstrued by
teachers as academic excellence, provides elite children with unfair advantages in securing
desirable socioeconomic outcomes. Importantly, this tradition assumes that the high-status
cultural signals are linked to a monolithic highbrow culture. However, every test of the
dimensionality of high culture finds not one, but rather two distinctive groupings: a books
and reading-related dimension and a beaux arts/ arts appreciation/ arts spectatorship

2
dimension. Building on this distinction, a growing body of evidence demonstrates that when
book-related forms of cultural resources are distinguished from other forms, the former and
not the latter account for much of educational success. This is because book-related
resources have a substantive link to academic-related skills such as vocabulary building,
counterfactual thinking, and cognitive flexibility, whereas highbrow arts consumption or
extracurricular activities have no substantive impact on academic skills. It is only when the
books and reading measures are (incorrectly) incorporated into a single "cultural capital"
measure with arts spectatorship that the signals as a group have an effect. In short, scholarly
culture is separate from arts spectatorship and, moreover, scholarly culture has a major
impact on educational performance and attainment, but arts spectatorship has little or no
effect. This is important because it undermines the claim that cultural resources are arbitrary
signals used by the elite to exclude or disadvantage others: instead, highbrow arts (the
arbitrary signals) are of little or no importance to education, while engagement in book-
related culture raises educational attainment.
The implication, which we shall examine here, is that scholarly culture endows children with
cognitive skills that intrinsically enhance academic performance, rather than scholarly culture
being merely an arbitrary signal of elite membership. If that implication is correct, we should
be able to detect effects of scholarly culture on cognitive skills – this is consistent with prior
research but has not previously been tested due to data limitations.
Another important aspect of the scholarly culture theory is its proposition that involvement in
books and reading benefits most disadvantaged children and not the children of the elite. This
contrasts with cultural reproduction theory and builds on the cultural mobility model, as
detailed below.
Our goals in this paper are to 1) validate the scholarly culture hypothesis in the context of
new outcomes that are essential to the claim of intrinsic connection between book-oriented
childhoods and educational and occupational success, namely, adult literacy, numeracy, and
solving problems using information and communication technologies (ICT); and 2) to
illuminate the inter- and intra-generational mechanisms through which scholarly culture
operates. We will also consider, more speculatively, if culture conceptualized thus is likely to
remain relevant in the era of digital literacy. In service of these goals, we present evidence
from 31 societies which participated in Rounds I and II of the Program for International
Assessment of Adult Competencies (PIAAC).

3
2. Cultural reproduction versus scholarly culture
2.1 Cultural reproduction
Cultural reproduction arguments, building on Bourdieu (1984, Bourdieu and Passeron, 1990),
suggest that elite parents strategize to equip their children with cultural signals which
teachers mistake for academic excellence and hence invest more in educating the elite
offspring (Goldthorpe, 2007, Jæger and Breen, 2016). Researchers followed Bourdieu in
distinguishing three states of so understood cultural capital, namely the embodied capital
which involves legitimate preferences and behaviours, the objectified capital such as books
and other transmittable physical goods and the institutionalised capital that involves
educational credentials and other indicators of class position (Lamont and Lareau, 1988). It is
commonly assumed that embodied capital is necessary for the objectified capital to be
mobilised and effectively used for enhancing one’s institutionalised capital. Empirical
research on cultural reproduction in the last 30 years has been based primarily on the
evidence from the United States (for a review of over 40 relevant studies see: Jæger and
Breen, 2016) and focused on the extent to which various forms of cultural capital in the
family of origin enhanced children’s educational achievement, and, in turn, educational
attainment. Virtually no research in the cultural reproduction tradition considered the
implications of cultural capital for adult outcomes other than educational attainment (but see:
Evans, et al., 2015).
Theoretically, cultural capital involves a range of elite status signals that is too broad to
operationalize (Lamont and Lareau, 1988), hence references to Bourdieu “wild” (Goldthorpe,
2007). In contrast, empirical research usually focuses on the participation in the highbrow
culture or Bourdieu’s “domesticated” capital (Goldthorpe, 2007), which involves educational
resources, cultural involvement, extracurricular activities, and reading climates, with
exposure to and possession of books treated merely as one of its many possible indicators
(Jæger and Breen, 2016, Kingston, 2001). Importantly, high status culture is seen a
monolithic, with all the aspects from reading to attendance at art galleries being
interchangeable indicators. Research in this tradition aims to demonstrate that educational
inequalities are maintained by elite parents who invest in cultural capital to prime their
offspring for socio-economic success regardless of their academic ability (Lareau and
Weininger, 2003). By contrast, the cultural mobility tradition, instigated by DiMaggio (1982),
rejects the logic of cultural reproduction theory arguing that it is the children from the lowest
socio-economic origins, and not the elite, who benefit most from endowment with cultural

4
resources. This tradition accords with the view that scholarly culture endows children with
cognitive skills, intellectual flexibility, and problem-solving capacity that endure throughout
their lives. These cognitive skills rather than arbitrary cultural signals of elite status are what
translates scholarly cultural endowment into educational and occupational success.
2.2 Scholarly culture
Scholarly culture theory highlights book-oriented socialization, indicated by adolescents'
home library size, as a source of cognitive competencies, skills and knowledge that are
valued not only in formal education (Evans, et al., 2014, Evans, et al., 2010) but also by
employers in different places and historical periods (Mateos-Romero and Salinas-Jiménez,
2016). Scholarly culture does not comprise arbitrary cultural signals that identify elite
members and earmark them for privileged positions in society: It enhances performance and
as such it is valued in various historical circumstances and by modest families as well as the
elite (Duchhardt, et al., 2015, Evans, et al., 2010).
2.2.1 Scholarly culture as social practice and way of life
While the exact processes through scholarly culture accrue may vary, they involve social
practices (Evans, et al., 2010, Reckwitz, 2002) in which books co-exist with specific mental
activities, the know-how and motivational states. Scholarly culture practice is like a “‘block’
whose existence necessarily depends on specific interconnectedness “of material and non-
material elements and which cannot be reduced to any one of these elements” (Reckwitz
2002: 250). This involves interactions with other household members but also solitary
activities with books, storytelling, imaginative play and vocabulary development. Parents
who encourage their children to read and enjoy books contribute to their educational success
through stimulating children’s cognitive skills (Kraaykamp and Notten, 2016, Park, 2008)
but, here, their behaviour is routine practice and not concerted cultivation (Lareau, 2011).
Books and objects such as e-readers, are not just disposable accessories to scholarly culture
practices but constitute their integral part and motive.
2.2.2 Scholarly culture enhances success in school
Both detailed studies of individual countries and studies with a broad international sweep find
that children who come from bookish homes get better grades and perform at a higher level
on standardized tests than children who are otherwise comparable on stratification and
demographic characteristics but come from families with smaller home libraries (Bodovski
and Farkas, 2008, Cheung and Andersen, 2003, Comber and Keeves, 1973, Heyneman and
Loxley, 1983, Jæger, 2011).

5
Research which sought to evaluate various aspects of home literacy environments in 25
societies found that household library size predicted school success already among fourth-
graders net of the reported literacy activities undertaken with parents and net of parental
attitudes to literacy (Park, 2008). Home libraries continue to facilitate students’ academic
achievement in adolescence (de Graaf, 1988, Evans, et al., 2014, Evans and Kelley, 2002).
The results on the standardized tests are particularly important because they are anonymously
graded, so there is no teacher to be duped by potentially arbitrary cultural signals.
2.2.3 Scholarly culture enhances educational attainment
With respect to educational gains, both detailed studies of individual countries and research
project with a global scope find that net of other influences, children from homes with larger
home libraries get more years of education overall and are also more likely to make all the
major educational transitions (Crook, 1997, de Graaf, et al., 2000, de Graaf, 1986, de Graaf,
1988, Evans, et al., 2010, Georg, 2004, Teachman, 1987). This has held since at least the
1940s, holds for societies across the whole span of socioeconomic development, held for
Communist societies as well as others, and holds for the disadvantaged groups for which it
has been assessed (Evans, et al., 2010).
2.2.4 Scholarly culture enhances occupational success
This topic received less attention, but the evidence to date is that, in the 27 societies studied
thus far, socialization into scholarly culture is associated with higher occupational attainment,
primarily indirectly by raising educational attainment, but also with an extra lift beyond that
(Evans, et al., 2015).
2.2.5 The effects are greatest for the most disadvantaged
As anticipated by the cultural mobility thesis, the effects of these cultural resources are
strongest for people from the most disadvantaged homes. Assessment of functional forms of
the effect of home library size on educational and occupational outcomes repeatedly shows
that a log linear form fits better than a linear form (Evans, et al., 2014, Evans, et al., 2015,
Evans, et al., 2010). The means that each additional book in a home library has much larger
benefits for families that only possess a few books than for families that already possess
many. This has been interpreted as the additional skills conferred per book: The first book
you read opens a whole new set of operational skills and cognitive possibilities; each
subsequent book continues to offer vocabulary expansion, new experiences in cognitive
complexity and intellectual flexibility, but at a reduced rate.

6
2.2.6 This paper's contribution: Scholarly culture confers cognitive skills that last into adulthood
It has long been argued that cognitive assets are closely entwined with life chances (Kohn
and Schooler, 1978, Spaeth, 1976). The specific contribution of scholarly culture theory is
that cognitive skills are the "missing link" heretofore unmeasured directly in studies of the
effects of book-oriented socialization on educational and occupational success. We cannot
here assess the specific mechanisms whereby book-oriented socialization instils cognitive
skills and intellectual flexibility, although we argue that it is through social practices in the
family which make up a bookish way of life (Reckwitz, 2002), rather than the more
instrumental concerted cultivation proposed in the cultural reproduction model (Lareau,
2011).
That model relegates the role of books to merely one of many possible indicators of parental
cultural capital in its objectified form, which is often individually owned and purchase-
related. However, book-oriented culture is a shared rather than individual resource. Thus, in a
twin study aimed at measuring individual capital, individual book ownership within twin
pairs correlated at 0.994 (Jæger and Møllegaard, 2017) which fits the concept of bookishness
as a shared family practice. Moreover, bookishness is more culturally acquired than purchase-
oriented (Lamont and Lareau, 1988). Books cost money but family bookshelves might be
filled with regular loans from libraries or bookish friends.
Building on this evidence, the current paper extends scholarly culture theory by
demonstrating that bookishness as social practice, to which youth are acculturated, creates
cognitive benefits which are not only immediate but also last into adulthood and are
independent of educational and occupational standing (although bookishness also
significantly enhances both forms of attainment).
It might be argued that the rapid rise of digital literacy calls into question the relevance of
bookish cultures for future generations, but we find no sign of a diminishing effect in more
recent cohorts of PIAAC participants. We take up this question more deeply in the
conclusion.
2.3 Hypotheses
Our hypotheses stem from an argument about broad benefits of bookish socialization. Hence,
we expect similar outcomes for literacy, numeracy and ICT skills:

7
Hypothesis 1: Diminishing returns with greatest gains at the bottom: Adult literacy,
numeracy and ICT problem solving skills develop more when home libraries grow
from tiny to medium than from large to enormous.
Hypothesis 2: Direct literacy benefits of growing up in scholarly culture: Exposure to larger
home libraries in adolescence enhances adult educational and occupational
attainment, but it also boosts literacy, numeracy and ICT skills net of parental
education or respondents’ attainment in adulthood. The impact on skill is contrary to
the cultural reproduction claim that cultural resources confer arbitrary signals.
Hypothesis 3: Life-long cumulative benefits of scholarly culture: larger home libraries in
adolescence benefit adult cognitive skills through stimulating more engagement in
out-of-work routine adulthood practices that maintain literacy, numeracy and ICT
skills but also irrespective of these activities.
3. Data, measurement and method
3.1 Data
The Programme for the International Assessment of Adult Competencies (PIAAC) measures
adult literacy, numeracy and information technology competencies (OECD, 2016). Data in
Rounds 1 and 2, used here, were collected between 2011 and 2015 (OECD, 2016) in person
or by telephone from representative samples of adults aged 16 through 65, regardless of
citizenship, nationality or language status. The survey was primarily a computer-based
assessment and only respondents with insufficient basic computer skills were given pen-and-
pencil tests. We use data from adults aged between 25 and 65 years from 31 societies:
Australia, Austria, Belgium, Canada, Chile, Czech Republic, Germany, Denmark, Spain,
Estonia, Finland, France, England and Northern Ireland (here referred to as the United
Kingdom), Greece, Ireland, Israel, Italy, Japan, Korea, Lithuania, Netherlands, Norway, New
Zealand, Poland, Russian Federation (with the exception of Moscow), Singapore, Slovak
Republic, Slovenia, Sweden, Turkey, and the United States. Response rates varied from 45 %
in Sweden to 81% in Turkey (for details see: OECD, 2016: Table 16.4). We had to exclude
Cyprus entirely, because it did not have variables related to everyday adult activities that
enhance literacy, numeracy or ICT skills. However, reduced models for Cyprus (available
upon request) align with the results we report here. Italy, France and Spain had no data on
digital problem-solving skills so, for this outcome, we had to limit our estimations to 28
societies.

8
3.2 Dependent variables
We consider all three domains of PIAAC assessment, namely 1) literacy, 2) numeracy and 3)
ICT problem solving. In PIAAC literacy is the ability to read effectively to participate in
society and achieve personal goals (OECD, 2016). Special emphasis was given to the
comprehension of digital texts displayed on screens that require navigation skills not needed
to read traditional print. Example items include interpreting preschool rules or user guides for
gym equipment. The literacy assessment did not include respondents’ writing skills. Rather,
test scores captured a range of basic through advanced comprehension skills, from reading
brief texts for a single piece of information to synthesising information from complex texts,
while making high-level inferences.
Numeracy in PIAAC is an ability to use mathematical concepts in everyday life. PIAAC
respondents answered questions gauging numeracy at different proficiency levels ranging
from simple sorting, arithmetic operations and recognising spatial representations in familiar
contexts to understanding abstract mathematical and statistical ideas presented in complex
contexts. Examples included interpreting graphical representations of trend data or science
problems involving measurement.
PIAAC also assessed the ability to use digital technology to communicate with others as well
as to gather, analyse and synthesise information. This domain involves a combination of
broadly understood computer literacy and ICT problem-solving skills. Three types of
problems were presented to respondents. The first involved evaluating the information
available on the Internet for quality and credibility, the second called for the application of
new technological tools such as spreadsheets for managing records, while the third involved
technical knowledge related to the operation of computer or navigating the Internet. These
skills are seen as “solving problems in technology-rich environments” (Desjardins, et al.,
2013, Kankaraš, et al., 2016). France, Italy and Spain did not provide any data on adult skills
in this domain and other societies had data for some respondents only. Hence our analysis of
this domain involves only 106,585 respondents in contrast to 162,955 respondents who have
data on literacy and numeracy (Supplementary materials Appendix Table 2 has descriptive
statistics).
PIAAC has been criticised for overemphasising generic skills used out of the meaningful
context of everyday lives as they may differ from the skills utilised in concrete social
situations (Tsatsaroni and Evans, 2014). Nevertheless, the survey provides the best up-to-date

9
comparative data on adult literacy, numeracy and technological problem-solving and as such
is optimal for our purposes. We use ten plausible values (OECD, 2016), standardized in the
pooled file in a manner that preserves their original variation.
3.3 Independent variables
Following prior studies of scholarly culture, we focus on the question about the number of
books respondents had at home in adolescence. All PIAAC respondents older than 16 were
asked: About how many books were there in your home when you were 16 years old? Do not
include magazines, newspapers or schoolbooks. To give an estimation, one metre of shelving
is about 40 books. Answer categories were: 10 books or less; between 11 and 25 books;
between 26 and 100 books; between 101 and 200 books; between 201 and 500; more than
500 books. We control for gender, age, parental education in years obtained by averaging
maternal and paternal education and respondent’s education in years (OECD, 2016).
Educational variables have been constructed based on PIAAC documentation for particular
societies. We also control for respondents’ occupational status in ISEI scores (Ganzeboom, et
al., 1992, Ganzeboom and Treiman, 2010). To ensure that the measure of home library in
adolescence does not reflect the variation in the intensity of adult activities, undertaken daily,
which maintain literacy, numeracy and technology skills, we use as control variables three
OECD indices (OECD, 2016). The first is an eight-item index of reading skills used at home,
that reflects reading basic and complex materials e.g. instructions, emails and professional
publications. The second is a six-item index of numeracy skills used at home that gauges the
frequency of calculating prices, budgets, using formulas, algebra or calculus. The third is a
six-item index of ICT skills used at home, which indicates how often respondents used email,
made purchases online, used spreadsheet software or participated in real-time discussions on
the Internet. Answer categories for all index items were: Never, Less than once a month, Less
than once a week but at least once a month, At least once a week but not every day and Every
day. Higher values of each index denote higher frequency (OECD, 2016). Prior to imputing
some missing values (Appendix Table 1), we standardised all independent variables except
for the number of books and respondents’ education as these are our key predictors for which
standardization obscures interpretability. Additionally, gender was left as a dummy variable.
Standardisation makes no difference to the substance of our results. In additional analyses,
we considered models which control for work-related activities (i.e. reading, numeracy and
ICT used at work) but their results are comparable to what we present here so we have opted

10
for more parsimonious models. Moreover, work-related variables are available only for sub-
samples of respondents.
3.4 Omitted variables and measurement error
We have no data on the types of books held in home libraries, although these matter, and
people in bookish households are more likely than others to read poetry, science,
mathematics or technology texts (de Graaf, 1986). Instead, we must focus on the boundaries
between people who surround themselves with many books and those who do not.
Cross-sectional data with retrospective indicators of scholarly culture are affected by
measurement error and indicators could reflect unmeasured influence of other confounding
variables which may lead to over-optimistic estimates (de Vries and de Graaf, 2008).
However, prior research using longitudinal data from Australia with corrections for
measurement error showed that a substantial impact of home library size on adult education
existed net of the effects of academic ability (IQ) of adults or their father’s scholarly habitus
(i.e. employment in occupations where use of books is common) as well as family income or
wealth (Evans, et al., 2010). Although we have no means of correcting for these potential
biases in ways available in longitudinal data, we argue that even somewhat optimistic
estimates are of value as an initial step in proposing and assessing hypotheses for future,
more stringent tests. To limit the potential for confounding adolescent and adult literacy-
enhancing activities, we control for adult daily out-of-work activities related to literacy,
numeracy and ICT skills that might correlate with scholarly culture experienced in
adolescence but are indicators of adult scholarly culture.
3.5 Method
We use ordinary least squares regressions with balanced repeated replicate (BRR) weights
that correct for the complex designs of PIAAC samples which vary from society to society
(Avvisati and Keslair, 2017, OECD, 2016). In pooled analysis we rely on senate weights
which adjust for complex sample design and ensure that each society contributes equally to
the analysis. To maximise the use of available information we imputed the missing data on
independent variables separately in each society (see Supplementary materials Appendix
Table 1), utilising chained equations and generating 10 imputed datasets to match the
number of plausible values (Royston, 2004). Most analyses, we present, comprised fitting ten
models, one for each plausible value, and combining the coefficients using the Rubin rule

11
(Little and Rubin, 1987). The substantive results are the same regardless of whether we use
the imputed data or listwise deletion of missing values.
For reporting purposes, we convert the number of books into mid-category values of
5, 18, 63, 151, 351 and 650 and use a natural log to capture the effect of books in one
coefficient. The value of 650 books was an arbitrary choice but other reasonable values make
no difference to our results. Sensitivity analyses showed that using natural logs is as effective
as using dummy variables in each country (see Supplementary materials Appendix Tables 3
through 5). Our full model for literacy is as follows:
Eq.1
Literacy
=
f(ln_Books_in_adolescence,
Parents’_education,
Respondent_education,
Respondent_occupation, Age,
Female,
Reading_at_home)
Our models predicting numeracy and ICT problem solving skills are as above, except
that we use the indices of numeracy activities at home and ICT activities at home,
respectively, in lieu of the reading at home index.
4. Results
4.1 Home libraries vary in size from society to society
The reported adolescent exposure to books at home varies considerably from society to
society (Table 1). Adults in Scandinavia and in some post-communist societies recall
growing up with large home libraries. The average library size in Norway was 212 and in
Sweden it was 210, in contrast to 192 in Denmark and 162 in Finland. Estonians grew up
with 218 books on average and the Czechs with 204. On the other end of the spectrum, the
average home library size in Turkey was 27 books, in Chile or Singapore 52 books, with the
average across the 31 societies at 115 books.
[Table 1]
Evidently, what counts as a large home library, varies considerably depending on societal
context. Nevertheless, these reports broadly correspond with reports in other data collections
in different time periods, where the residents of post-Soviet and Scandinavian societies also
reported larger home libraries than the denizens of English speaking parts of the world

12
(Evans, et al., 2015, Evans, et al., 2010). Local standards of bookishness matter for local
practices and success at school, therefore our key conclusions are based on society-specific
analyses with the pooled analyses serving merely as a convenient summary of what we
establish, society-by-society, as nearly universal effects.
4.2 Greatest gains at the bottom: exposure to books in adolescence enhances adult literacy,
numeracy and ICT skills
We begin from comparisons of unadjusted literacy, numeracy and problem-solving scores
obtained by respondents who grew up in households with relatively few as opposed to many
books (Figure 1a).
[Figure 1]
Growing up with almost no books is associated with literacy levels at about half a standard
deviation below the mean in the pooled sample. Having had approximately 80 books in
adolescent home library raises literacy levels to the average while from about 350 books
onwards further growth in the library size is not associated with significant literacy gains.
The relationship is remarkably similar for numeracy and ICT skills (Figure 1b and 1c) but
gains in ICT skills, associated with larger library sizes, are not as steep. While it may appear
that bookishness matters less for digital competencies, the important caveat of the data about
ICT skills, however, is that they are available only from a PIAAC subsample, which, on
average, was two years younger than the entire sample, had more educated parents and more
books in their adolescent libraries (152 versus 127). Thus, it is only to be expected that for
this group of respondents, given the general pattern of smaller returns in the upper end of the
distribution, the benefits will be more moderate. Notwithstanding that, the log-linear
relationship between the number of books and the dependent variables holds for all three
cognitive domains and the natural log of the average number of books represents well these
relationships, when compared to a dummy variable specification, which we also considered
(see Supplementary materials Appendix Tables 3 through 5). Therefore, we settled on using
the natural log of books as our key independent variable.
4.3 Books in adolescence enhance adult literacy, numeracy and digital problem-solving:
total and direct effects
In the pooled data the exposure to home libraries in adolescence explains about 17% of
variance in adult literacy, 17% of variance in numeracy and about 8% of variance the ICT
skills in the younger and better educated subsample (Model 1: Total effects in Table 2). The

13
books coefficient suggests that adult literacy raises by 0.274 of a standard deviation which
means that when a library grows from 18 to 63 books, the literacy gain is about 0.34 of a
standard deviation but when it grows from 350 to 650 books, the associated gain is only 0.17.
The corresponding coefficients for numeracy and ICT skills are also sizeable, at 0.277 and
0.207, respectively, and imply the diminishing returns pattern stipulated by our first
hypothesis, while contradicting the expectation that the elite benefit most from books as
signals of cultural capital.
[Table 2]
Model 2 in Table 2 summarizes the direct effects and shows that home library size remains a
significant predictor of adult literacy even after we control for adult education (each year of
which raises adult literacy by 0.079 of a standard deviation), occupation, parental education,
sex, age and routine reading that adults undertake outside of work. The effects of predictors
other than books and respondents’ education can be compared as they have been standardized
to a common metric. While education is the strongest predictor (as shown later in path
analyses), high occupational status and frequent home reading also enhance literacy. Women
fall marginally behind men’s literacy test scores; net of their socio-demographic
characteristics. The literacy skills of older people are also somewhat weaker. The key
conclusion here is that exposure to larger home libraries in adolescence has a positive direct
effect on literacy, numeracy and ICT skills even when educational and occupational
attainment of adults are controlled for. While much of the benefit is via attainment some of it
is attainment-independent, in line with Hypothesis 2.
Auxiliary analyses, not shown in Table 2, demonstrate that in one third of PIAAC societies a
negative interaction exists between respondents’ education and the number of books in
adolescence when literacy and numeracy are modelled as outcomes. This suggests that, in
some locations, growing up with large libraries makes a particularly large contribution to
adult competencies of people with lower educational credentials which further supports the
logic of cultural mobility arguments when they are applied to intra-generational mobility.
The analysis presented here is, to our knowledge, the first to consider how home libraries
matter for numeracy and the impact shown is no less than for literacy (but see: Braun, 2018),
and it shows the net boost at 0.132 of a standard deviation. Other effects resemble the effects

14
of independent variables in the literacy model except for sex, as women lag behind men more
in numeracy than literacy.
Books in adolescence enhance also adult ICT skills, which are generally stronger amongst the
better educated and those in higher status occupations. The younger respondents who
frequently undertake ICT activities outside of work, have the strongest ICT skills. A one-
standard deviation growth in the index of ICT activity at home rises technological problem-
solving skills by 0.170 of a standard deviation.
Overall, the direct effects in Table 2 are consistent with Hypothesis 3 which expected
cumulative benefits of exposure to scholarly culture. Adolescent engagement with books at
home enhances adult competencies not only via attainments and regular reading habits in
adulthood (as shown later in our path analyses which estimate not only effects on literacy, but
also adult education, occupation and reading at home) but also directly, as a lifelong
propensity to routinely include books into one’s cultural and material environment. Table 2
suggests that about 40% of the effect of books is not explained by socio-economic attainment
or demographic factors, and even if this estimate is optimistic, its direction and nearly
universal relevance (seen later in our society-specific results) supports the view that when
bookishness becomes an integral part of early socialisation, it later enhances a specific suite
of cognitive skills, through fostering the practice of surrounding oneself with books. This
practice pays off in formal education in adolescence and it continues to benefit adults later in
life in ways inherent to but also entirely independent of attainment-related returns.
4.4 Bookish adolescents with lower secondary education credentials become as literate,
numerate and technologically apt in adulthood as university graduates who grew up with
only a few books
Predicted values, or average partial estimates (Williams, 2012) based on Table 2, are in
Figure 2 and illustrate the relative importance of scholarly culture and formal education for
the cognitive outcomes. In the PIAAC cohort, people who were between 25 and 65 years of
age between 2011 and 2015, grew up without any books, and managed to finish only lower
secondary school (9 years) typically performed in the literacy test at about -0.55 of a standard
deviation below the mean (Figure 2a). Their counterparts with university degrees had roughly
average literacy levels (0.00). The same level of literacy was achieved by people who were
surrounded by many books in adolescence but whose schooling ended in Year 9 (0.02). So,
literacy-wise, bookish adolescence makes up for a good deal of educational advantage.
Adults in their early forties (the average age in our PIAAC sample is nearly 45 years) from

15
bookish homes who completed university had, unsurprisingly, highest levels of literacy, at,
roughly, half a standard deviation above the mean (0.57) with all other predictors kept at
mean values. For gender we used the average proportion of women which was 51%, other
means are in Appendix Table 2 in Supplementary materials.
[Figure 2]
The relative advantages of home libraries for adult numeracy mirror those for literacy (Figure
2b). Respondents with 9 years of education who grew up without books had numeracy levels
less than half a standard deviation below the mean (-0.59). Their bookish counterparts had
average numeracy skills (0.03) which is close to the expected value for university-educated
respondents who grew up with very few books (-0.01). Thus, adolescent exposure to books
compensates for shortcomings not only in adult literacy but also numeracy, which are
equivalent to additional years of education.
The same pattern, albeit with less variation due to the restricted sample, holds also for the
ICT skills (Figure 2c). Exposure to books in adolescence boosts technological competencies
by over 0.4 of a standard deviation. This gain is similar for respondents with lower secondary
and university education. What varies is the typical level of these skills in different
respondent groups. Lower secondary education and no books in adolescence are associated
with below average (-0.48) technology skills (this prediction is for people around 43 years of
age, the values would be even lower for an older cohort). Lower secondary graduates who
had plenty of books in their home, sport ICT skills comparable to those of their bookless but
university-educated peers (-0.06 versus -0.14 in Figure 2c).
Overall, the impact of book-oriented socialisation is substantial. The details for each of the 31
societies are in Tables 3, 4 and 5 and these society-specific analyses are our main evidence.
While some variation in society-specific predictors of literacy occurs (Table 3), the results on
pooled data in Table 2 approximate well society-specific patterns. The total effects of home
library size on literacy are large everywhere and the direct effect is insignificant only in
Lithuania, where it is explained by occupational attainment (not shown in Table 3). As
expected, respondents’ education, occupational status and reading activities at home are
strong predictors of superior literacy nearly everywhere, but respondents clearly benefit from
adolescent exposure to books above and beyond these effects.
[Table 3] [Table 4] [Table 5]

16
Table 4, which shows society-by-society estimates for numeracy, also lends credence to the
message in the pooled model: both total and direct effects of home library sizes are sizeable
in each society without exception. Finally, bookishness directly enhances ICT skills in many
societies even in the younger and more socially select sub-sample (Model 2 in Table 5),
although the impact is not as steep as for literacy and numeracy. This holds for all societies
except Greece and Israel where the ICT benefits of growing up with books are explained
entirely by respondents’ occupational attainment.
4.5 How do adolescent libraries matter? Path analyses: Literacy as an example
Adolescent bookishness influences adult outcomes in many ways. The fully standardized
coefficients from path analyses, estimated in Mplus 7 (Muthén and Muthén, 2013), show that
adolescent home library size significantly enhances not only literacy, but also adult
education, occupation and non-work routine reading activities (Figure 3). Because the
patterns are similar for all three cognitive domains, the results for numeracy and ICT skills
are in Supplementary materials (Appendix Figures 1 and 2).
Indirectly, books benefit literacy mostly through educational, but also through occupational
attainment as well as a boost from routine reading practices undertaken by adults. Yet, the
direct effect of books on literacy remains non-trivial at 0.171. The coefficient depicting the
enhancement of adult reading habits by adolescent exposure to books is 0.158, with the
greatest benefit occurring for educational attainment 0.323 in contrast to 0.068 for
occupational status. Here, the reading habits indicator is a predictor of literacy because it
gauges the frequency of activities that precede literacy tests taken by PIAAC respondents. In
sum, the benefits of bookishness for attainment are beyond question. The direct effects of
books on adult literacy and reading habits fit in with the scholarly culture argument but not
with theories of elite closure that are central to traditional cultural capital arguments. Early
exposure to books in parental home matters because books are an integral part of routines and
practices which enhance life-long cognitive competencies.
5 Conclusion
5.1 Summary
Our results show that adolescent exposure to books is an integral part of social practices that
foster long term cognitive competencies spanning literacy, numeracy and ICT skills, as
anticipated by Hypothesis 2. These competencies facilitate educational and occupational

17
attainment, but they also lay a foundation for life-long routine activities that enhance literacy
and numeracy, irrespective of attainment, as anticipated by Hypothesis 3. Our findings from
31 societies surveyed between 2011 and 2015 support the scholarly culture argument and
stand in opposition to the theory of cultural closure/ cultural reproduction in which cultural
capital has no relationship to concrete skills and, instead, functions as a signal that legitimises
the social exclusion of non-elite persons. These results are robust to sensitivity tests and they
also corroborate the expectation from scholarly culture theory that the first few books make
the greatest contribution to cognitive skills of all three kinds, as Hypothesis 1 predicted. In
other words, home library size has a loglinear effect on cognitive, numerical, and problem-
solving skills that endure throughout life.
Of course, our data are not perfect: they have only a single indicator (rather than multiple
indicators) of scholarly culture; retrospective data are not perfectly reliable, and our key
indicator could incorporate unmeasured heterogeneity in family backgrounds. These issues
all call for replication of this analysis with data that address these deficiencies.
5.2 Discussion: The future of scholarly culture
We must consider the possibility that as knowledge societies move towards digital literacy
and numeracy, the consumption of printed materials and books will become obsolete as an
indicator of scholarly culture. For now, however, the beneficial effects of home libraries in
adolescence are large and hold in many different societies with no sign of diminution over
time. Moreover, home library size is positively related to higher levels of digital literacy so,
the evidence suggests that for some time to come, engagement with material objects of
scholarly culture in parental homes, i.e. books, will continue to confer significant benefits for
adult ICT competencies. In the US, over 80% of readers who used e-books 2011 already read
many print books which they supplemented with digital reading (Zhang and Kudva, 2014).
Indeed, recent research indicates that reading printed rather than digital texts enhances
comprehension and retention of information (Mangen, et al., 2013). Specifically, metanalyses
of studies, that compare reading of print and digital material, point to the advantages of
reading books for deeper comprehension of complex content (Singer and Alexander, 2017),
retrieving specific information (Mangen, et al., 2013), and facilitating shared family reading
time (Kucirkova and Littleton, 2016).
Therefore, future surveys should include not only questions about the possession of printed
books and e-readers, as was done in Programme for International Student Assessment 2015,

18
but also collect information about the use of audio books, e-books, and printed books,
distinguishing also their type of content. For the time being, however, the perception that
social practice of print book consumption is passé is premature.
5.3 Discussion: How does scholarly culture confer cognitive skills?
Now that we have established that scholarly culture as indicated by the size of home libraries,
confers enduring cognitive skills in literacy, numeracy, and technology, the next burning
question becomes: "How does this come about?"
We will propose some possibilities for future research on engagement with digital and printed
books. Role modelling: Children emulate parents who read (de Graaf, 1986, de Graaf, 1988).
Acquisition of specific strategies proposed by significant others or discovered in books
themselves: children build "toolkits" of strategies that they apply in multiple situations
(Swidler, 1986). Stimulation of cognitive skills through family social practices: books are
interwoven with positive affect, specific mental activities, know-how, and motivational states
(Reckwitz, 2002). Storytelling, imaginative play, charades, and vocabulary development
come to mind (Evans, et al., 2010). We suggest that scholarly culture is a way of life rather
than concerted cultivation (Lareau, 2011).
6. References
Avvisati, F., Keslair, F., 2017. REPEST: Stata module to run estimations with weighted replicate
samples and plausible values. Statistical Software Components.
Bodovski, K., Farkas, G., 2008. “Concerted cultivation” and unequal achievement in elementary
school. Social Science Research 37, 903-919. doi:10.1016/j.ssresearch.2008.02.007
Bourdieu, P., 1984. Distinction: A social critique of the judgment of taste. Harvard University Press,
Cambridge, MA.
Bourdieu, P., Passeron, J.-C., 1990. Reproduction in education, society and culture. Sage.
Braun, H., 2018. How long is the shadow? The relationships of family background to selected adult
outcomes: results from PIAAC. Large-scale Assessments in Education 6, 4.
Cheung, S.Y., Andersen, R., 2003. Time to read: family resources and educational outcomes in
Britain. Journal of Comparative Family Studies 34, 413-433.
Comber, L., Keeves, J.P., 1973. Science education in nineteen countries: an empirical study. Wiley,
New York.
Crook, C.J., 1997. Cultural practices and socioeconomic attainment: the Australian experience.
Greenwood Publishing Group, Westport, CT.
de Graaf, N.D., de Graaf, P.M., Kraaykamp, G., 2000. Parental cultural capital and educational
attainment in the Netherlands: a refinement of the cultural capital perspective. Sociology of
Education 73, 92-111. doi:10.2307/2673239
de Graaf, P.M., 1986. The impact of financial and cultural resources on educational attainment the
Netherlands. Sociology of Education 59, 237-246. doi:10.2307/2112350
de Graaf, P.M., 1988. Parents' financial and cultural resources, grades, and transition to secondary
school in the Federal Republic of Germany. European Sociological Review 4, 209-221.
doi:10.1093/oxfordjournals.esr.a036485

19
de Vries, J., de Graaf, P.M., 2008. Is the intergenerational transmission of high cultural activities
biased by the retrospective measurement of parental high cultural activities? Social Indicators
Research 85, 311-327. doi:10.1007/s11205-007-9096-4
Desjardins, R., Thorn, W., Schleicher, A., Quintini, G., Pellizzari, M., Kis, V., Chung, J., 2013.
OECD skills outlook 2013: First results from the Survey of Adult Skills.
DiMaggio, P., 1982. Cultural capital and school success: the impact of status culture participation on
the grades of U.S. high school students. American Sociological Review 47, 189-201.
doi:10.2307/2094962
Duchhardt, C., Jordan, A.K., Ehmke, T., 2015. Adults’ use of mathematics and its influence on
mathematical competence. International Journal of Science and Mathematics Education.
doi:10.1007/s10763-015-9670-1
Evans, M., Kelley, J., Sikora, J., 2014. Scholarly culture and academic performance in 42 nations.
Social Forces 92, 1573-1605. doi:doi.org/10.1093/sf/sou030
Evans, M., Kelley, J., Sikora, J., Treiman, D.J., 2015. Scholarly culture and occupational success in
31 societies. Comparative Sociology 14, 1-43.
Evans, M.D.R., Kelley, J., 2002. Cultural resources and educational success: The beaux arts vs
scholarly culture. In: Evans, M. D. R., and Kelley, J., (Eds.), Australian Economy and Society
2001: Education, Work, and Welfare. Federation Press, Sydney, pp. 54-65.
Evans, M.D.R., Kelley, J., Sikora, J., Treiman, D.J., 2010. Scholarly culture and educational success
in 27 nations. Research in Social Stratification and Mobility 28, 171-197.
Ganzeboom, H.B.G., de Graaf, P.M., Treiman, D.J., 1992. A standard international socio-economic
index of occupational status. Social Science Research 21, 1-56. doi:10.1016/0049-
089X(92)90017-B
Ganzeboom, H.B.G., Treiman, D.J., 2010. Occupational status measures for the new international
standard classification of occupations ISCO-08; with a discussion of the new classification,
http://www.harryganzeboom.nl/isol/isol2010c2-ganzeboom.pdf.
Georg, W., 2004. Cultural capital and social inequality in the life course. European Sociological
Review 20, 333-344. doi:10.1093/esr/jch028
Goldthorpe, J.H., 2007. "Cultural capital": some critical observations. Sociologica 2, 1-22.
doi:10.2383/24755
Heyneman, S.P., Loxley, W.A., 1983. The effect of primary-school quality on academic achievement
across twenty-nine high-and low-income countries. American Journal of Sociology 88, 1162-
1194.
Jæger, M.M., 2011. Does cultural capital really affect academic achievement? New evidence from
combined sibling and panel data. Sociology of Education 84, 281-298.
doi:10.1177/0038040711417010
Jæger, M.M., Breen, R., 2016. A dynamic model of cultural reproduction. American Journal of
Sociology 121, 1079-1115.
Jæger, M.M., Møllegaard, S., 2017. Cultural capital, teacher bias, and educational success: New
evidence from monozygotic twins. Social Science Research 65, 130-144.
doi:10.1016/j.ssresearch.2017.04.003
Kankaraš, M., Montt, G., Paccagnella, M., Quintini, G., Thorn, W., 2016. Skills matter: further results
from the Survey of Adult Skills. OECD skills studies, OECD Publishing, Paris.
Kingston, P.W., 2001. The unfulfilled promise of cultural capital theory. Sociology of Education 74,
88-99. doi:10.2307/2673255
Kohn, M.L., Schooler, C., 1978. The reciprocal effects of the substantive complexity of work and
intellectual flexibility: A longitudinal assessment. American Journal of Sociology 84, 24-52.
Kraaykamp, G., Notten, N., 2016. Parental cultural socialization and educational attainment. Trend
effects of traditional cultural capital and media involvement. Research in Social Stratification
and Mobility 45, 63-71. doi:10.1016/j.rssm.2016.08.003
Kucirkova, N., Littleton, K., 2016. The digital reading habits of children: A national survey of
parents’ perceptions of and practices in relation to children’s reading for pleasure with print
and digital books. Book Trust, London.
Lamont, M., Lareau, A., 1988. Cultural capital: allusions, gaps and glissandos in recent theoretical
developments. Sociological Theory 6, 153-168. doi:10.2307/202113

20
Lareau, A., 2011. Unequal childhoods: Class, race, and family life. Univ of California Press,
Berkeley.
Lareau, A., Weininger, E.B., 2003. Cultural capital in educational research: A critical assessment.
Theory and Society 32, 567-606. doi:10.1023/B:RYSO.0000004951.04408.b0
Little, R.J.A., Rubin, D.B., 1987. Statistical analysis with missing data. Wiley, New York.
Mangen, A., Walgermo, B.R., Brønnick, K., 2013. Reading linear texts on paper versus computer
screen: Effects on reading comprehension. International Journal of Educational Research 58,
61-68.
Mateos-Romero, L., Salinas-Jiménez, M.D.M., 2016. Skills heterogeneity among graduate workers:
real and apparent overeducation in the Spanish labor market. Social Indicators Research, 1-
18. doi:10.1007/s11205-016-1338-x
Muthén, L.K., Muthén, B.O., 2013. Mplus user's guide. Seventh edition. . Muthén & Muthén, Los
Angeles, CA.
OECD, 2016. Technical report of the survey of adult skills (PIAAC) (2nd edition). OECD, Paris.
Park, H., 2008. Home literacy environments and children's reading performance: a comparative study
of 25 countries. Educational Research and Evaluation 14, 489-505.
doi:10.1080/13803610802576734
Reckwitz, A., 2002. Toward a theory of social practices: a development in culturalist theorizing.
European Journal of Social Theory 5, 243-263. doi:10.1177/13684310222225432
Royston, P., 2004. Multiple imputation of missing values. The Stata Journal 4, 227-241.
Singer, L.M., Alexander, P.A., 2017. Reading on paper and digitally: what the past decades of
empirical research reveal. Review of Educational Research, 87, 1007-1041
doi:10.3102/0034654317722961.
Spaeth, J.L., 1976. Cognitive complexity: a dimension underlying the socioeconomic achievement
process. In: Sewell, W., Hauser, R. M., and Featherman, D., (Eds.), Schooling and
Achievement in American Society. Academic Press, New York, pp. 103-131.
Swidler, A., 1986. Culture in action - symbols and strategies. American Sociological Review 51, 273-
286.
Teachman, J.D., 1987. Family background, educational resources, and educational attainment.
American Sociological Review 52, 548-557.
Tsatsaroni, A., Evans, J., 2014. Adult numeracy and the totally pedagogised society: PIAAC and other
international surveys in the context of global educational policy on lifelong learning.
Educational Studies in Mathematics 87, 167-186. doi:10.1007/s10649-013-9470-x
Williams, R., 2012. Using the margins command to estimate and interpret adjusted predictions and
marginal effects. Stata Journal 12, 308-331.
Zhang, Y., Kudva, S., 2014. E‐books versus print books: Readers' choices and preferences across
contexts. Journal of the Association for Information Science and Technology 65, 1695-1706.

21
Table 1. Home library size in adolescence: percentages and means for 31 societies, PIAAC 2011-2015
Number of books
Around 5
Around 20
Around 65
Around 150
Around 350
500 +
Mean
Std.
Dev.
N
%
%
%
%
%
%
Australia
13
13
35
18
13
7
148
178
6465
Austria
15
18
33
15
11
7
131
173
4133
Belgium
26
21
30
11
9
4
95
145
4051
Canada
18
16
33
16
12
5
125
163
22059
Chile
46
23
21
6
3
2
52
110
4136
Czech Republic
2
6
33
27
22
10
204
186
4602
Denmark
9
12
29
21
19
11
192
201
6221
Estonia
4
9
28
23
22
13
218
205
6240
Finland
9
14
34
18
16
8
162
183
4569
France
22
17
31
14
9
6
117
167
5811
Germany
12
15
34
17
14
8
151
180
4316
Greece
37
23
26
7
5
2
62
114
4235
Ireland
23
18
31
13
10
4
107
154
5210
Israel
17
14
29
18
14
9
153
187
4009
Italy
32
23
26
11
5
3
75
127
4064
Japan
18
20
36
12
9
4
102
146
4403
Korea
23
19
35
13
7
3
91
134
5584
Lithuania
15
23
35
14
9
5
109
151
4369
Netherlands
16
15
29
16
15
8
154
188
4196
New Zealand
12
12
33
18
16
9
166
190
4804
Norway
6
9
30
22
20
14
212
208
3982
Poland
16
20
35
15
10
4
111
149
4892
Russian Federation
7
15
35
22
14
7
154
174
2653
Singapore
40
23
27
6
3
1
52
98
4319
Slovak Republic
11
18
38
19
10
4
117
143
4566
Slovenia
24
21
33
13
7
4
92
139
4457
Spain
22
21
33
12
9
4
102
152
4979
Sweden
8
9
29
21
21
13
210
207
3627
Turkey
60
21
13
4
1
0
27
62
4335
United Kingdom
15
16
33
16
13
7
143
179
7585
United States
20
18
32
15
9
5
114
159
4083
Pooled data
21
18
32
15
10
5
115
160
162955

22
Figure 1. Adult numeracy, literacy and ICT problem solving skills by home library size in
adolescence: pooled data
Data: PIAAC 2011-2015

23
Table 2. OLS models with BRR weights predicting adult literacy, numeracy, and ICT problem-solving.
Pooled data for 31 societies, PIAAC 2011-2015
Literacy
Numeracy
ICT skills
Coeff.
S.E.
Coeff.
S.E.
Coeff.
S.E.
Model 1: Total effects
Books ln
0.274**
0.005
0.277**
0.004
0.207**
0.009
Constant
-1.046**
0.021
-1.059**
0.019
-0.859**
0.037
R-squared
0.17
0.17
0.08
Model 2: Direct effects
Books ln
-0.116**
0.007
-0.129**
0.005
0.091**
0.009
Education years
-0.079**
0.003
-0.083**
0.003
0.052**
0.005
Parental education
-0.060**
0.008
-0.025**
0.008
0.060**
0.009
Occupation ISEI
-0.100**
0.009
-0.119**
0.008
0.148**
0.011
Female
-0.042**
0.014
-0.223**
0.014
-0.120**
0.017
Age
-0.050**
0.007
-0.032**
0.007
-0.184**
0.012
Reading at home
-0.085**
0.008
Numeracy at home
-0.096**
0.007
ICT use at home
0.170**
0.014
Constant
-1.405**
0.038
-1.412**
0.037
-1.114**
0.073
R-squared
0.32
0.34
0.25
N of respondents
162,955
162,955
106,585
N of societies
31
31
28
**Coefficients statistically significant at p=0.01
Note: All independent variables standardized to a common metric except books in natural logs and education in years

24
Figure 2. Predicted literacy, numeracy and ICT problem solving gains for respondents with different
educational attainment who grew up in homes with different library sizes, based on Table 2
Figure 2 continued next page

25

26
Table 3. Adult literacy regressed on home library size in adolescence: total and direct effects
Model 1: Total effects
Model 2: Direct effects
Literacy
Books ln
Books ln
Education
Parental
education
Occupation
Reading at
home
Female
Age
Coef.
S.E.
R2
Coef.
S.E.
Coef.
S.E.
Coef.
S.E.
Coef.
S.E.
Coef.
S.E.
Coef.
S.E.
Coef.
S.E.
R2
N
Australia
0.27**
0.01
0.14
0.14**
0.01
0.09**
0.01
-0.013
0.017
0.19**
0.02
0.19**
0.02
-0.09**
0.03
-0.10**
0.02
0.32
6465
Austria
0.26**
0.01
0.17
0.11**
0.01
0.05**
0.01
0.062**
0.019
0.19**
0.02
0.14**
0.02
-0.06
0.03
-0.14**
0.01
0.31
4133
Belgium
0.22**
0.01
0.12
0.06**
0.01
0.10**
0.01
0.068**
0.015
0.16**
0.02
0.14**
0.02
-0.11**
0.03
-0.14**
0.02
0.38
4051
Canada
0.26**
0.01
0.13
0.12**
0.01
0.09**
0.01
0.021
0.012
0.20**
0.02
0.19**
0.01
-0.12**
0.03
-0.10**
0.01
0.33
22059
Chile
0.30**
0.02
0.15
0.07**
0.03
0.12**
0.01
0.001
0.045
0.12**
0.03
0.07**
0.02
-0.17**
0.05
-0.15**
0.03
0.41
4136
Czech Republic
0.23**
0.02
0.09
0.09**
0.02
0.08**
0.01
0.001
0.032
0.10**
0.03
0.14**
0.03
-0.04
0.04
-0.12**
0.02
0.27
4602
Denmark
0.26**
0.01
0.13
0.11**
0.01
0.07**
0.01
0.028**
0.014
0.17**
0.02
0.23**
0.02
-0.04
0.02
-0.16**
0.01
0.34
6221
Estonia
0.25**
0.01
0.12
0.11**
0.01
0.08**
0.01
0.019
0.012
0.09**
0.01
0.10**
0.02
-0.05
0.03
-0.10**
0.01
0.25
6240
Finland
0.28**
0.02
0.13
0.09**
0.02
0.07**
0.01
0.007
0.024
0.18**
0.02
0.20**
0.03
0.02
0.04
-0.25**
0.02
0.34
4569
France
0.28**
0.01
0.17
0.09**
0.01
0.09**
0.00
0.036**
0.016
0.14**
0.01
0.15**
0.02
-0.02
0.02
-0.13**
0.01
0.38
5811
Germany
0.30**
0.01
0.18
0.11**
0.01
0.09**
0.01
0.031
0.021
0.15**
0.02
0.19**
0.02
-0.06**
0.03
-0.16**
0.02
0.38
4316
Greece
0.19**
0.01
0.08
0.08**
0.02
0.05**
0.01
0.082**
0.027
0.05
0.03
0.06**
0.02
0.11**
0.04
0.08**
0.03
0.15
4235
Ireland
0.25**
0.01
0.15
0.10**
0.01
0.09**
0.01
0.026
0.020
0.11**
0.02
0.13**
0.02
-0.11**
0.03
-0.02
0.02
0.31
5210
Israel
0.27**
0.01
0.12
0.08**
0.01
0.09**
0.01
0.078**
0.018
0.22**
0.02
0.13**
0.02
-0.13**
0.03
-0.22**
0.02
0.35
4009
Italy
0.25**
0.01
0.15
0.10**
0.01
0.07**
0.01
-0.044
0.033
0.06**
0.02
0.09**
0.02
0.01
0.04
-0.04
0.02
0.27
4064
Japan
0.20**
0.01
0.11
0.08**
0.01
0.10**
0.01
-0.011
0.016
0.08**
0.02
0.11**
0.02
0.01
0.03
-0.22**
0.02
0.33
4403
Korea
0.23**
0.01
0.15
0.06**
0.01
0.07**
0.00
0.026
0.015
0.10**
0.02
0.08**
0.01
-0.08**
0.02
-0.15**
0.01
0.36
5584
Lithuania
0.16**
0.01
0.06
0.02
0.02
0.04**
0.01
0.073**
0.019
0.09**
0.02
0.16**
0.03
0.03
0.03
-0.04**
0.02
0.19
4369
Netherlands
0.28**
0.01
0.18
0.12**
0.01
0.10**
0.01
0.002
0.018
0.19**
0.02
0.15**
0.03
-0.10**
0.03
-0.20**
0.02
0.41
4196
New Zealand
0.26**
0.01
0.14
0.15**
0.01
0.08**
0.01
0.022
0.015
0.23**
0.02
0.12**
0.02
-0.05
0.03
-0.09**
0.02
0.32
4804
Norway
0.30**
0.01
0.16
0.15**
0.01
0.06**
0.01
0.039**
0.016
0.22**
0.02
0.18**
0.03
-0.08**
0.03
-0.13**
0.02
0.32
3982
Poland
0.27**
0.01
0.15
0.08**
0.02
0.08**
0.01
0.037
0.025
0.11**
0.02
0.09**
0.02
0.03
0.03
-0.06**
0.02
0.29
4892
Russian Federation
0.14**
0.02
0.04
0.11**
0.03
0.01
0.01
0.023
0.027
-0.02
0.03
0.13**
0.04
0.06
0.04
0.06**
0.02
0.08
2653
Singapore
0.33**
0.01
0.13
0.04**
0.01
0.16**
0.01
-0.001
0.020
0.22**
0.02
0.15**
0.02
-0.06**
0.03
-0.16**
0.02
0.53
4319
Slovak Republic
0.25**
0.01
0.16
0.11**
0.01
0.07**
0.01
0.071**
0.020
0.01
0.02
0.12**
0.02
0.01
0.02
-0.02
0.01
0.25
4566
Slovenia
0.25**
0.01
0.13
0.05**
0.01
0.13**
0.01
0.097**
0.016
0.12**
0.02
0.11**
0.02
-0.06**
0.03
-0.11**
0.02
0.33
4457
Spain
0.31**
0.01
0.20
0.11**
0.01
0.09**
0.01
-0.015
0.020
0.12**
0.02
0.11**
0.02
-0.11**
0.03
-0.11**
0.01
0.37
4979
Sweden
0.35**
0.01
0.20
0.20**
0.02
0.08**
0.01
-0.007
0.016
0.20**
0.02
0.16**
0.03
-0.10**
0.04
-0.14**
0.02
0.35
3627
Turkey
0.27**
0.02
0.11
0.10**
0.02
0.08**
0.01
-0.019
0.041
0.03
0.02
0.07**
0.02
-0.14**
0.05
-0.09**
0.03
0.25
4335
United Kingdom
0.26**
0.01
0.15
0.15**
0.01
0.05**
0.01
0.057**
0.022
0.25**
0.02
0.16**
0.02
-0.07**
0.03
-0.05**
0.02
0.29
7585
United States
0.29**
0.02
0.18
0.10**
0.01
0.10**
0.01
0.128**
0.021
0.18**
0.02
0.08**
0.02
-0.05
0.03
-0.05**
0.01
0.39
4083
**Coefficients statistically significant at p=0.01, constants and their SE not shown due to space constraints but available from the authors upon request

27
Table 4. Adult numeracy regressed on home library size in adolescence: total and direct effects
Model 1: Total effects
Model 2: Direct effects
Numeracy
Books ln
Books
ln
Education
Parental
education
Occupation
Numeracy at
home
Female
Age
Coef.
S.E.
R2
Coef.
S.E.
Coef.
S.E.
Coef.
S.E.
Coef.
S.E.
Coef.
S.E.
Coef.
S.E.
Coef.
S.E.
R2
N
Australia
0.27**
0.01
0.13
0.14**
0.01
0.08**
0.01
0.002
0.017
0.20**
0.02
0.21**
0.02
-0.29**
0.03
-0.06**
0.02
0.35
6465
Austria
0.24**
0.01
0.15
0.11**
0.01
0.07**
0.01
0.039
0.022
0.19**
0.02
0.16**
0.02
-0.19**
0.03
-0.07**
0.01
0.31
4133
Belgium
0.20**
0.01
0.10
0.05**
0.01
0.10**
0.01
0.049**
0.016
0.15**
0.02
0.13**
0.02
-0.28**
0.03
-0.11**
0.02
0.39
4051
Canada
0.24**
0.01
0.12
0.11**
0.01
0.10**
0.01
0.032**
0.013
0.19**
0.01
0.19**
0.01
-0.31**
0.02
-0.07**
0.01
0.35
22059
Chile
0.33**
0.03
0.17
0.08**
0.02
0.13**
0.01
-0.004
0.035
0.16**
0.03
0.13**
0.03
-0.38**
0.04
-0.14**
0.02
0.52
4136
Czech Republic
0.27**
0.02
0.13
0.13**
0.02
0.09**
0.01
0.014
0.026
0.11**
0.03
0.13**
0.03
-0.16**
0.04
-0.09**
0.02
0.34
4602
Denmark
0.23**
0.01
0.11
0.10**
0.01
0.08**
0.01
0.014
0.014
0.16**
0.02
0.18**
0.02
-0.21**
0.03
-0.07**
0.01
0.32
6221
Estonia
0.24**
0.01
0.12
0.09**
0.01
0.09**
0.01
0.031**
0.012
0.11**
0.01
0.15**
0.01
-0.18**
0.02
-0.05**
0.01
0.31
6240
Finland
0.24**
0.01
0.11
0.08**
0.01
0.06**
0.01
0.006
0.022
0.19**
0.02
0.23**
0.02
-0.19**
0.03
-0.15**
0.02
0.34
4569
France
0.29**
0.01
0.18
0.09**
0.01
0.10**
0.01
0.024
0.017
0.19**
0.01
0.18**
0.01
-0.20**
0.02
-0.07**
0.01
0.45
5811
Germany
0.29**
0.01
0.17
0.10**
0.01
0.11**
0.01
0.025
0.020
0.14**
0.03
0.21**
0.02
-0.23**
0.03
-0.11**
0.01
0.43
4316
Greece
0.19**
0.01
0.09
0.06**
0.02
0.07**
0.01
0.045
0.026
0.09**
0.03
0.07**
0.02
-0.11**
0.04
0.06**
0.02
0.22
4235
Ireland
0.25**
0.01
0.14
0.10**
0.01
0.09**
0.01
0.029
0.020
0.13**
0.02
0.13**
0.02
-0.27**
0.03
-0.01
0.02
0.32
5210
Israel
0.27**
0.01
0.12
0.08**
0.02
0.10**
0.01
0.077**
0.022
0.21**
0.02
0.16**
0.02
-0.31**
0.04
-0.14**
0.02
0.35
4009
Italy
0.26**
0.01
0.17
0.13**
0.01
0.07**
0.01
-0.086**
0.029
0.08**
0.03
0.10**
0.02
-0.23**
0.03
-0.05**
0.02
0.31
4064
Japan
0.19**
0.01
0.11
0.08**
0.01
0.10**
0.01
-0.007
0.018
0.12**
0.02
0.12**
0.01
-0.17**
0.03
-0.11**
0.01
0.31
4403
Korea
0.23**
0.01
0.15
0.06**
0.01
0.09**
0.01
0.020
0.015
0.11**
0.01
0.08**
0.01
-0.14**
0.02
-0.12**
0.01
0.38
5584
Lithuania
0.23**
0.01
0.12
0.07**
0.01
0.06**
0.01
0.092**
0.018
0.14**
0.02
0.15**
0.02
-0.08**
0.03
-0.06**
0.02
0.29
4369
Netherlands
0.25**
0.01
0.16
0.11**
0.01
0.10**
0.01
-0.014
0.016
0.16**
0.02
0.15**
0.01
-0.24**
0.02
-0.10**
0.01
0.41
4196
New Zealand
0.28**
0.01
0.15
0.17**
0.01
0.08**
0.01
0.019
0.014
0.23**
0.02
0.17**
0.02
-0.27**
0.03
-0.05**
0.02
0.36
4804
Norway
0.30**
0.01
0.15
0.16**
0.02
0.08**
0.01
0.019
0.018
0.20**
0.02
0.16**
0.02
-0.24**
0.03
-0.05**
0.02
0.33
3982
Poland
0.26**
0.01
0.15
0.09**
0.01
0.08**
0.01
0.037
0.026
0.09**
0.02
0.16**
0.02
-0.13**
0.03
-0.02
0.02
0.29
4892
Russian Federation
0.13**
0.02
0.05
0.09**
0.03
0.03**
0.01
0.006
0.022
0.01
0.03
0.12**
0.03
0.01
0.04
0.03
0.02
0.09
2653
Singapore
0.34**
0.01
0.13
0.05**
0.01
0.19**
0.01
-0.008
0.018
0.26**
0.02
0.12**
0.02
-0.16**
0.03
-0.13**
0.02
0.58
4319
Slovak Republic
0.30**
0.01
0.19
0.14**
0.01
0.08**
0.01
0.067**
0.022
0.05**
0.02
0.16**
0.01
-0.08**
0.02
0.01
0.01
0.33
4566
Slovenia
0.27**
0.01
0.15
0.07**
0.01
0.16**
0.01
0.085**
0.022
0.12**
0.02
0.13**
0.02
-0.19**
0.03
-0.11**
0.02
0.37
4457
Spain
0.30**
0.01
0.21
0.12**
0.01
0.08**
0.01
-0.031
0.019
0.12**
0.02
0.11**
0.01
-0.26**
0.03
-0.09**
0.01
0.41
4979
Sweden
0.32**
0.01
0.18
0.18**
0.01
0.08**
0.01
-0.015
0.017
0.20**
0.02
0.17**
0.02
-0.26**
0.03
-0.06**
0.02
0.34
3627
Turkey
0.33**
0.02
0.12
0.12**
0.02
0.10**
0.01
-0.008
0.049
0.08**
0.03
0.09**
0.02
-0.39**
0.05
-0.11**
0.04
0.35
4335
United Kingdom
0.26**
0.01
0.14
0.15**
0.01
0.04**
0.01
0.055**
0.020
0.27**
0.02
0.18**
0.02
-0.28**
0.03
0.01
0.02
0.32
7585
United States
0.31**
0.02
0.19
0.11**
0.01
0.11**
0.01
0.121**
0.020
0.17**
0.02
0.17**
0.02
-0.30**
0.03
-0.02
0.01
0.43
4083
**Coefficients statistically significant at p=0.01, constants and their SE not shown due to space constraints but available from the authors upon request

28
Table 5. Adult information and communication technology problem solving skills regressed on home library size in adolescence: total and direct effects
Model 1: Total effects
Model 2: Direct effects
ICT skills
Books ln
Books ln
Education
Parental
education
Occupation
ICT at home
Female
Age
Coef.
S.E.
R2
Coef.
S.E.
Coef.
S.E.
Coef.
S.E.
Coef.
S.E.
Coef.
S.E.
Coef.
S.E.
Coef.
S.E.
R2
N
Australia
0.20**
0.01
0.09
0.13**
0.01
0.05**
0.01
-0.01
0.02
0.16**
0.02
0.17**
0.02
-0.07
0.04
-0.16**
0.02
0.27
4775
Austria
0.21**
0.01
0.11
0.11**
0.02
0.04**
0.01
0.04
0.02
0.13**
0.02
0.17**
0.02
-0.19**
0.03
-0.27**
0.02
0.32
3023
Belgium
0.16**
0.01
0.06
0.04**
0.01
0.07**
0.01
0.06**
0.02
0.14**
0.02
0.25**
0.02
-0.14**
0.03
-0.30**
0.02
0.39
3283
Canada
0.21**
0.01
0.08
0.10**
0.01
0.06**
0.01
0.04**
0.01
0.17**
0.02
0.23**
0.02
-0.10**
0.03
-0.19**
0.01
0.27
17026
Chile
0.26**
0.02
0.11
0.07**
0.02
0.09**
0.02
0.09**
0.04
0.12**
0.05
0.17**
0.03
-0.18**
0.06
-0.31**
0.04
0.38
2399
Czech Republic
0.27**
0.03
0.07
0.11**
0.03
0.05**
0.01
0.11**
0.04
0.18**
0.03
0.16**
0.03
-0.10
0.05
-0.24**
0.02
0.27
3269
Denmark
0.23**
0.01
0.10
0.10**
0.01
0.06**
0.01
0.02
0.02
0.15**
0.01
0.21**
0.02
-0.13**
0.03
-0.33**
0.02
0.38
5123
Estonia
0.27**
0.01
0.10
0.11**
0.01
0.06**
0.01
0.07**
0.01
0.13**
0.02
0.19**
0.02
-0.11**
0.04
-0.27**
0.02
0.33
3963
Finland
0.25**
0.01
0.10
0.07**
0.01
0.06**
0.01
0.04
0.02
0.15**
0.02
0.24**
0.03
-0.08**
0.03
-0.39**
0.02
0.44
3647
Germany
0.26**
0.02
0.12
0.10**
0.02
0.06**
0.01
0.03
0.02
0.18**
0.03
0.24**
0.02
-0.12**
0.04
-0.27**
0.02
0.37
3510
Greece
0.15**
0.02
0.04
-0.01
0.03
0.07**
0.01
0.04
0.04
0.08**
0.04
0.22**
0.04
0.05
0.05
-0.07**
0.03
0.18
2557
Ireland
0.24**
0.01
0.13
0.11**
0.01
0.08**
0.01
0.03**
0.01
0.10**
0.02
0.18**
0.02
-0.16**
0.03
-0.19**
0.02
0.35
3430
Israel
0.17**
0.02
0.04
0.02
0.02
0.05**
0.01
0.11**
0.03
0.20**
0.04
0.26**
0.03
-0.12**
0.05
-0.26**
0.02
0.27
2564
Japan
0.15**
0.02
0.04
0.06**
0.02
0.08**
0.01
0.03
0.02
0.11**
0.03
0.19**
0.03
-0.19**
0.04
-0.40**
0.03
0.30
2720
Korea
0.16**
0.01
0.06
0.04**
0.01
0.06**
0.01
0.04**
0.02
0.10**
0.02
0.11**
0.01
-0.17**
0.03
-0.30**
0.03
0.29
3533
Lithuania
0.24**
0.02
0.10
0.05**
0.02
0.04**
0.01
0.10**
0.02
0.12**
0.03
0.29**
0.03
-0.05
0.04
-0.14**
0.02
0.37
3018
Netherlands
0.22**
0.01
0.12
0.09**
0.01
0.07**
0.01
0.01
0.02
0.15**
0.02
0.23**
0.02
-0.14**
0.03
-0.26**
0.02
0.39
3699
New Zealand
0.24**
0.01
0.10
0.15**
0.01
0.05**
0.01
0.03
0.02
0.21**
0.02
0.19**
0.02
-0.03
0.04
-0.23**
0.02
0.29
4301
Norway
0.28**
0.01
0.15
0.14**
0.01
0.07**
0.01
0.01
0.02
0.16**
0.02
0.19**
0.02
-0.13**
0.03
-0.30**
0.02
0.40
3424
Poland
0.24**
0.03
0.07
0.07**
0.03
0.07**
0.02
0.10**
0.04
0.15**
0.04
0.11**
0.04
-0.25**
0.06
-0.23**
0.03
0.24
2446
Russian Federation
0.14**
0.04
0.02
0.07
0.05
0.01
0.01
-0.02
0.04
0.02
0.05
0.21**
0.04
0.08
0.10
-0.03
0.06
0.08
1737
Singapore
0.20**
0.01
0.07
0.05**
0.01
0.12**
0.01
0.01
0.02
0.16**
0.02
0.12**
0.02
-0.09**
0.03
-0.31**
0.02
0.39
3046
Slovak Republic
0.19**
0.02
0.06
0.09**
0.02
0.06**
0.01
0.07**
0.03
0.04
0.03
0.04
0.02
-0.07
0.04
-0.09**
0.02
0.14
2560
Slovenia
0.28**
0.02
0.12
0.09**
0.02
0.12**
0.02
0.12**
0.02
0.16**
0.03
0.18**
0.03
-0.14**
0.03
-0.28**
0.02
0.37
3176
Sweden
0.32**
0.02
0.16
0.15**
0.01
0.07**
0.01
-0.01
0.02
0.19**
0.02
0.22**
0.02
-0.13**
0.03
-0.31**
0.02
0.42
3155
Turkey
0.18**
0.02
0.06
0.04
0.02
0.05**
0.01
0.15**
0.05
0.04
0.04
0.15**
0.04
0.06
0.07
-0.10**
0.04
0.17
1681
United Kingdom
0.23**
0.01
0.11
0.12**
0.02
0.04**
0.01
0.05**
0.02
0.23**
0.02
0.22**
0.02
-0.22**
0.03
-0.19**
0.02
0.36
6166
United States
0.22**
0.02
0.10
0.08**
0.02
0.06**
0.01
0.09**
0.02
0.17**
0.02
0.20**
0.02
-0.11**
0.04
-0.14**
0.02
0.31
3354
**Coefficients statistically significant at p=0.01, constants and their SE not shown due to space constraints but available from the authors upon request
Note: Data not available in France, Italy and Spain

29
Figure 3. Impact of adolescent home library size on adult education, occupation, frequency of reading outside of work and literacy skills. Standardized solution STDYX,
generated in Mplus 7. Unstandardized solution for the effects of books corresponds to what is shown in Table 2.
Note: Only significant pathways shown (correlations between independent variables not shown). Path analyses for numeracy and ICT skills are in Supplementary materials
Legend
books: home library size in adolescence
par_educ: parental education
educat: respondent’s education
readhome: index of reading at home
occupat: occupational status ISEI score